Skip to content

Commit a89b932

Browse files
committed
fixed issue with sci notation on yaml
1 parent 1003540 commit a89b932

1 file changed

Lines changed: 1 addition & 1 deletion

File tree

configs/generator/sst2_hf.yaml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -31,7 +31,7 @@ training:
3131
# torch_dtype: "auto" # (optional) lets HF pick fastest dtype
3232

3333
# ── optimiser & scheduler ─────────────────────────────────────────────────
34-
learning_rate: 5e-5 # good starting LR for GPT‑2 on small corpora
34+
learning_rate: 0.00005 # good starting LR for GPT‑2 on small corpora
3535
warmup_ratio: 0.1
3636

3737
# ── misc performance knobs ────────────────────────────────────────────────

0 commit comments

Comments
 (0)